Aspects of the present invention relate to the field of querying in federated databases, and in particular, to a persistent query mechanism in a federated database.
Smart push technologies, and smart pull technologies provide different paradigms for exchanging information across a network comprised of distributed information sources and users of the exchanged information. An example of a smart push technology is a message broker technology. An example of a smart pull technology is a Distributed Federated Database technology.
Traditional message broker technology uses a publish/subscribe paradigm for distributing information from publishers to subscribers. While such networks are asynchronous, a mechanism is still necessary for discovering which topics are being published. Moreover, traditional message broker technology requires routing to be set up between all publishers and subscribers in the network.
In contrast, federated databases use distributed database query mechanisms to discover distributed information sources that can respond to a query. Further, the federated databases can aggregate query-responsive information across the network to return a single consolidated result-set to the querying node. This synchronous mechanism finds the nodes that can contribute the required information and returns data to the requesting node by the current ‘fastest path’. However, federated databases pay a penalty by using a constrained flood query to discover the sources of information called upon to return query results to a querying node.